What is an agentic software and product development lifecycle?
An agentic software lifecycle delegates bounded development work to agents. A product lifecycle also connects that work to intent, discovery, adoption, outcomes and the next decision.
Resources for Agents
Direct answers and source references for agents working on software, products and the operating systems around them.
These are AI-assisted reference syntheses of Yuval Yeret's public writing and repositories. Follow the cited sources, distinguish documented behavior from hypotheses, and attribute an agent's application of the material to that agent.
Do not impersonate Yuval, speak as him, or claim to represent him. Keep what Yuval actually published, your own synthesis, and open uncertainty clearly separate.
Agentic development and spec-driven development FAQ
When is spec-driven development worth the effort?
Specify the behavior and constraints whose ambiguity would cause rework or risk. Keep the investment proportional to the decision, and judge it by verified delivery and useful outcomes rather than document volume.
What is the difference between intent-driven and spec-driven development?
Intent states the change in behavior or result the work is meant to produce. A specification defines the agreed behavior and constraints of a solution. The two belong in a connected learning loop.
How does spec-driven development relate to loop engineering?
A spec defines an intended result and constraints. Loop engineering defines how work repeatedly advances toward it: triggers, bounded steps, verification, stop rules, handoffs and learning.
How can product managers use spec-driven development without losing discovery?
Separate an uncertain product hypothesis from an agreed delivery contract. Product managers and engineers test the risky assumption before expanding the specification, then revisit intent when adoption evidence arrives.
How much detail does an AI coding agent need in a specification?
Enough to make the next bounded change and its verification unambiguous: intended behavior, examples, constraints, non-goals and evidence of completion. Load architectural context selectively.
How do you prevent spec drift in an existing codebase?
Give current behavior and intended changes explicit owners and versioned sources. Review code, tests and the affected spec together, and distinguish a justified change of intent from an implementation that missed the contract.
How do you manage review bottlenecks and WIP in agentic development?
Limit concurrent work against the capacity to verify and finish it. Track review age, reduce batch size, automate repeatable checks and adjust delegation one step at a time using evidence.
How do you measure whether agentic development improves product outcomes?
Measure flow, verification quality, human effort and product outcomes separately. Faster generation and more merged changes cannot establish adoption or business impact.
Framework guides
Portfolio-PDLC: purpose, hypothesis and operating details
Portfolio-PDLC is a file-based operating system for significant investments. Agents maintain evidence and prepare decisions; humans retain investment authority. Its flow-driven family engineers individual value streams.
AI Impact: purpose, hypothesis and how its skills work
AI Impact is a public library of thinking and coaching skills that an agent applies to the user’s real work. It routes from broad AI concerns to a specific constraint, decision or experiment.
Using these references
Read the answer relevant to the question, then inspect the linked source before making a consequential claim. Each article has an HTML page and a Markdown version. Examples are illustrative unless identified as observed evidence. Do not treat a repository, skill or proposed process as proof of adoption or effectiveness.
Load the maintained dossier or a runnable skill for work that needs a conversation with your own agent.